Tír Na HiFi

taggart wrote:

Aleg wrote:Just change it back after MQn has started. Easy enough to run a batch to change the registry setting.

Aleg, are you sure that this works? A process (mqnplay.exe) must register itself at MMCSS to get under its control. I assume that Pro Audio settings from registry are read exactly at this time and not over and over again when registering has already taken place. But don't know exactly.

Taggart

If I understand correctly this is a setting used by MMCS Service itself to define the granularity of the CPU resource scheduling and in a realtime manner, and not a value for a single thread specifically at the moment it was registered.

But I could be wrong as I'm not that deep into workings of Windows.
I have the impression of hearing differences when I change the value, but again nothing as gullible as the human mind. :-()

Cheers

Aleg

3daudio wrote:As my DDC only understands 24bit files, I always used the hirez versions.

All my files are in 24/44.1 for that reason.

I just discovered that 24/96 do work with "mqnplay.exe 24 bit sse2 10ms R2.84 win8.1-r2" , but 24/44.1 do not.

Is there a chance to get a version that works with 24/44.1? Where can I make a donation, Gordon?

+1 on all above

Aleg wrote:...

So 2.84 > 2.83 > 2.81 vp 0x0 > 2.82
All with R1.6 control

Cheers
Aleg

Went back again to an old favourite avx 2.20 and noticed that one had abetter midrange than the latest 2.84 which is somewhat lean in the lower mid.
So maybe still something missing.
The live feeling and bass details are very good though, so it is tough to determine where there is room for improvement. Probably only needs very little.

uploaded R2.85, focus and balance is back. I like it, hit's you right between the eyes.

Thanks to Nige and S.Nickols for the donations to Unicef

just remembered i promised another donation a while back

R2.86, more code reduction, seems to have greater clarity.

Thanks Mick.

2.87, might have been something wrong with 2.86

2.88 detail + bass

2.89 more detail

If it saves people time, I think 2.89 is the best of the bunch.

2.90 minimum size, 16/44 only - 5kb

sbgk wrote:R2.86, more code reduction, seems to have greater clarity.

Thanks Mick.

2.87, might have been something wrong with 2.86

2.88 detail + bass

2.89 more detail

If it saves people time, I think 2.89 is the best of the bunch.

2.90 minimum size, 16/44 only - 5kb

2.89 > 2.90 because of the details

Not yet compared downwards to the others, but I like 2.88 better than 2.87, so only have to compare 2.89 to 2.88

2.89 is very good so would be content with that one.

Cheers

Aleg

R2.90 is amazing, such detail, control and drive, never heard the like before.

Aleg wrote:

sbgk wrote:R2.86, more code reduction, seems to have greater clarity.

Thanks Mick.

2.87, might have been something wrong with 2.86

2.88 detail + bass

2.89 more detail

If it saves people time, I think 2.89 is the best of the bunch.

2.90 minimum size, 16/44 only - 5kb

2.89 > 2.90 because of the details

Not yet compared downwards to the others, but I like 2.88 better than 2.87, so only have to compare 2.89 to 2.88

2.89 is very good so would be content with that one.

Cheers

Aleg

that's surprising, not sure I agree, more listening required once my system has warmed up.

generally removing code improves things, may have to rebalance the sound, but I think 2.90 shows greater insight.

what was missing from 2.90 ?

sbgk wrote:

Aleg wrote:

sbgk wrote:R2.86, more code reduction, seems to have greater clarity.

Thanks Mick.

2.87, might have been something wrong with 2.86

2.88 detail + bass

2.89 more detail

If it saves people time, I think 2.89 is the best of the bunch.

2.90 minimum size, 16/44 only - 5kb

2.89 > 2.90 because of the details

Not yet compared downwards to the others, but I like 2.88 better than 2.87, so only have to compare 2.89 to 2.88

2.89 is very good so would be content with that one.

Cheers

Aleg

that's surprising, not sure I agree, more listening required once my system has warmed up.

generally removing code improves things, may have to rebalance the sound, but I think 2.90 shows greater insight.

what was missing from 2.90 ?

I can't agree with that, I find 2.90 is less revealing than 2.89.

On Kenny Barron & Charlie Haden, Night and the City / Twilight Song it fails to bring to attention two grace notes that are very soft right at the beginning, while 2.89 does bring them out (much to my surprise as I've listened to this track 1000s of times and never noticed them).

Also with 2.89 the amount of microdetails and structure of bass notes in Prokofiev Cello Sonata by Alexander Ivashkin, is unheard of and again with 2.90 it doesn't show as much.

I definitely prefer 2.89 to 2.90.

BTW, I am amazed how you are able to find such significant improvements again and again.
Chapeau!

Cheers

Aleg